You know the old business saying: in any project, the best you can hope for is to achieve two out of three from the set {Quick, Cheap, Good}? As in, "Sure, we can lay off half the programming team and ship by the end of the month, but don't come crying about the reviews", or "Yeah, we can deliver a flawless product in 25% less time than our best-case estimate, if you are willing to hire this $500/hr programmer to finish the engine"?

It's a classic problem in resource allocation: you can't possibly do everything that you might want to do in an ideal world, so you have to prioritize.

If your MBA-speak detector has just gone DefCon 4, relax: resource allocation is not only the producer's job. In fact, several of the most fundamental design decisions in any game depend on how you decide to spend your most precious asset: the user's brain activity. A game with Quake III Arena's break-neck pace and Civilization II's world-building scope might seem way cool in theory, but I suspect that it would overwhelm most mid-level deities, never mind the newbie mainstream buyer your company's president desperately wants to attract. So, what do you do?

1. The Joys of Short-Term Memory
Psychologists estimate that human short-term memory is equipped to manipulate a maximum of about 7 unrelated concepts at a time;  the more alien the ideas, the fewer can be managed safely. However, once an object, an idea or a technique has been properly internalized (or a set of concepts have been merged into a "gestalt"), it becomes second nature and no longer requires as much effort to manipulate; otherwise, we'd all still be struggling with poorly-tied shoelaces.

For example, consider the plight of the newbie who picks up a first-person shooter for the first time. There are no graphical user interface elements to guide him (and rightly so, as we'll discuss later), and the dozens of keyboard/mouse command combos he must learn to cope with are so arcane that he is likely to spend at least 10-15 minutes running into walls, firing bazookas at inoffensive artwork, and switching to a pocket knife before charging at the Terminator. Manipulating the basic commands is all he can handle; better make that training level VERY easy, and introduce the first monster in the third or fourth room only. On the other hand, a guy like Thresh may only need one of his seven short-term memory "slots" to handle "first-person interface" as a whole, which allows him to concentrate on more advanced features from the get go.

Your job as game designer is to make sure that your player is never overwhelmed by an inordinately large number of new concepts to master at the same time. On the other hand, new gameplay elements should be introduced into your game regularly (say, one or two per major level), otherwise your smartest players will get bored. (Come to think of it, this is yet another form of resource allocation, isn't it?)

Let us define these concepts in game terms:

A "depth-first" game, Half-Life for example, focuses on a narrow set of user experiences explored in great detail;

A "breadth-first" game (say, Civilization II) covers an enormous variety of topics but in a highly abstracted way.

In the rest of this article, we will use these definitions to identify which gameplay, artificial intelligence and user interface devices are most appropriate to particular games and genres.

First-person games
A first-person game (or its second cousin, the "trailing camera" third-person platformer) is the ultimate in depth-first entertainment. Your player is intimately linked to a single "specialist" character, for example a thief or a commando soldier, and should be able to do/feel/understand absolutely everything that this character can. The game consists of making the user exploit the character's talents and counteract his shortcomings.

In a game like this, sensory immersion is key. The best user interface is no interface at all: a computer screen is already quite limited in the range of vision it can provide, so avoid cluttering it with buttons and menu bars. Use as many graphic, audio, tactile (and soon olfactory) stimuli as you can to plunge your player into the character's world.

In all likelihood, your first-person game will be based on pure emotion; appropriate lighting effects and music, cunning enemies with whom the player can establish one-to-one relationships, realistic imagery and suspense are your most important tools. On a related note, make sure that you provide the player with enough sensory information to make the illusion work. I remember an utterly unplayable 3D pinball game where the player experienced the world from the point of view of the ball; it looked pretty, but the player's flippers were usually out of the camera's field of view. Three guesses as to how that worked, and the first two don't count.

Squad-level games
This is a game where the player controls a small party (usually 3-6 characters), either as their commander or as a "possessing force" taking over each one in turn. Baldur's Gate is the quintessential squad-game, with one possible twist: usually, in a squad game, it doesn't matter which character(s) survive(s), as long as someone accomplishes the mission, while BG requires the original player character to reach the end of the game for storyline purposes.

In a game like this, each non-player character in the user's team must be as detailed and unique as possible, and the individual-level AI controlling them when the player isn't must be very powerful to create suspension of disbelief. (If you ever release a game in which the wizard becomes catatonic as soon as the player takes control of the paladin, I will give your email address to the Dark Lord.) The player will get to "know" each of his virtual teammates personally, but the level of personal commitment involved will be somewhat lower than in a FPS: either you control one character and the others are mere assistants/tools, or you "possess" each one in turn and become part-owner of all and sole-owner of none.

Most of the time, an isometric view of some sort is still your best bet in squad-level games, because it combines a fair sense of immediacy with the possibility to watch your entire team at work at a glance. Only if the player is expected to control the same character all the time is a first-person view truly appropriate: otherwise, changes in "possession" will create disorientation, and they are extremely hard to pull off well in the real-time setting that first-person view really demands.

Baldur's Gate adopts a turn-based combat mode, which "feels right" in an isometric system (and would be out of place in a first-person game), which allows the player to control each character in as much detail as he wants.

As a rule, the squad-game player is like a platoon sergeant or a team leader: "one of the guys", whose job is to make the unit smoother by taking on some of the key tasks himself. If all of the NPC's in your squad game are identical, you may be missing out on something special.

Village-level games
The next step on the depth/breadth ladder is the village-level game, where you control a relatively high number of abstracted units instead of individual characters. Most real-time strategy games fall within this category: in Age of Empires, a "villager" unit represents dozens of people, they are all interchangeable, you do not have to move their arms to make them pick up each individual berry, and only the most psychotic of players will ever develop an emotional attachment to one of them. Team sports games fit the bill as well.

Village games require formation-level artificial intelligence, in which characters (each of which probably has a smaller and simpler set of actions than in the squad-level game described above) must be able to function within coordinated teams, or at least avoid stepping on each other's toes. They also demand some form of high-level point of view instead of (or in addition to) the more immersive 3D work needed above. There is a reason why generals direct battles from hilltops and maps instead of mucking in the trenches: they need the global, abstracted representation of the situation to be able to work.

The village-game player should be treated like a field captain or construction company manager: he must have a reasonable understanding of everything that's going on, but should not be expected to do it all himself.

World-level games
This is the all-out strategy game, à la Civilization or Star Trek: Birth of the Federation. The player commands a very large team/army/empire through a set of abstract actions with or without real-world counterparts.

In a game like this, there is no gameplay-related requirement for realistic imagery (witness chess and checkers). A 2D overhead view works perfectly well, and serves to reinforce the emotional distance which must be implemented between player and units. This is a game of intellect; if your player feels pangs of guilt every time one of his frigates is blown up in Master of Orion, his life will be a sorry thing indeed.

The variety of topics covered within a single world-level game can be staggering, which forces a breadth-first implementation: nobody will argue that Master of Orion's economics are realistic, but they get the job done and keep the game rolling. Civilization would be unplayable if the user had to manipulate each chariot in combat or handle farmers' markets directly. On the other hand, Birth of the Federation may dive a little too deep into the micro-management of parts of the game. For example, even in highly advanced galaxies (tech levels 8 or above), new colonies are established at tech level 1 and 2; upgrading them to the point where they can become significant contributors requires a LONG time (hours of game time) and regular attention on the player's part, because only one upgrade to a particular item can be on a colony's work list at any given time. The player must either repeat orders time and time again, or let the colony fester at its Neo-Neanderthal level.

In a world-level game, the player should act like a five-star general or CEO, making the wide-ranging decisions and leaving the implementation details to AI vice-presidents and colonels. Otherwise, the game will crawl to its death. Remember the old board wargames which took 6 hours to setup and three weeks to play? They had a limited audience, precisely for this reason.

Finally, because of the sheer number of parameters within the player's control in a world-level game, gaining access to each one can become a serious user-interface issue. Avoid multiple layers of menus and buttons whenever possible; right-clicking the mouse anywhere in Birth of the Federation invokes a handy shortcut menu providing quick access to any of the game's tactical screens, which eliminates much of the tedious navigation which might be required to support all of the game's features.

3. Pacing According to Scope
A final note: make sure that your game is paced in a way that is suitable to the amount of work required from the user.

Remember the fast-good-cheap dilemma? If your player controls a single character with a well-defined skill set, you can throw as many monsters and traps at him as you want; that's what he expects, and also what he needs to stay focused. He doesn't have to think, and all of his brainpower can be devoted to reaction.

However, if the game requires the player to plan ahead or act with precision, give him enough time to do so. Real-time strategy games have a limited number of units on screen and only a few resources to collect for this very reason: add more, and the player will be overwhelmed. Similarly, a game like Master of Orion could never work in real-time, because there is simply too much to do for a single person!

Conclusion
Some players easily get drawn into fast-paced, emotion-filled games. Others prefer the more sedate, thinking games. And yet another group likes both. Real-time strategy is a successful blending of the two; whether other attempts at the same symbiosis (like the RTS/first-person "possession" combination seen in Dungeon Keeper) will carve their own permanent niches remains to be seen.

In the meantime, knowing how to streamline your game to provide the user with exactly the experience he needs without burdening him with unwanted micro- (or macro-) management chores will make you a more effective designer. Now get back to work!

Bio
François-Dominic Laramée is a freelance interactive game designer, developer and producer.  He has been involved in one form or another of the game industry, whether PC, console, online, set-top box or even play-by-mail, for the past decade, including more than 8 years of experience as Head of Studio, Game Designer, Software Engineer, Producer and Quality Assurance Manager in the interactive entertainment and spoken dialogue interfaces industries.  Learn more on his website at: http://pages.infinit.net/idjy/ 

GIGnews is a publication of GIGnews.com, Inc.
"Get In the Game" is a registered trademark used with permission.
© 1999- 2005 GIGnews.com, Inc.
Legal